Chapter 5: Reliability

Reliability, in a broad statistical sense, is synonymous with:
1. consistently good
2. consistently bad
3. consistency
4. validity
A reliability coefficient is:
1. an index
2. a proportion of the total variance attributed to true variance
3. unaffected by a systematic source of error
4. all are correct
Which of the following is true of systematic error? Systematic error:
1. significantly lowers the reliability of a measure
2. insignificantly lowers the reliability of a measure
3. increases the reliability of a measure
4. has no effect on the reliability of a measure
As the degree of reliability increases, the proportion of:
1. total variance attributed to true variance increases
2. total variance attributed to true variance decreases
3. total variance attributed to error variance increases
4. none are correct
Computer-scorable items have virtually eliminated error variance, because of:
1. item sampling
2. scorer differences
3. content sampling
4. test-takers' reactions to environmental variables
Christopher Titus did not read and study Chapter 1 in the text until the night before an exam was given over Chapters 1-5. Christopher amazed himself (and others) by getting 90% of the items on the exam correct. As a budding expert in tests and measurement, you might best explain Christopher's success with reference to:
1. the concept of item sampling, especially as it relates to whoever wrote the exam
2. the concept of error, especially as it relates to whoever scored Christopher's exam
3. the concept of norming, especially as it relates to members of the normative sample
4. all are correct
Poorly worded items that cause students to differentially respond to the same questions contribute to what type of error variance?
1. content sampling
2. test administration error
3. test-scoring and interpretation variance
4. content sampling and test-scoring and interpretation variance
What type of reliability estimate is obtained by correlating pairs of scores from the same person (or people) on two different administrations of the same test?
1. parallel forms
2. split-half
3. test-retest
4. none are correct
What type of reliability estimate would be appropriate only when evaluating the reliability of a test that measures a trait that is relatively stable over time?
1. parallel forms
2. split-half
3. alternate forms
4. test-retest
Test-retest reliability estimates would be least appropriate for:
1. IQ tests
2. tests that measure moment-to-moment mood(s)
3. academic achievement tests on topics, such as ancient history
4. tests that measure art aptitude
An estimate of test-retest reliability is often referred to as a coefficient of stability when the time interval between the test and retest is more than:
1. 30 days
2. 60 days
3. 3 months
4. 6 months
Which of the following is true for parallel forms of a test?
1. the means of the observed scores are equal for the two forms
2. the variances of the observed scores are equal for the two forms
3. the means and variances of the observed scores are equal for the two forms
4. none are correct
Which source of error variance affects parallel- or alternate forms reliability estimates, but does not effect test-retest estimates?
1. fatigue
2. learning
3. practice
4. item sampling
Test-retest estimates of reliability are referred to as measures of ____________, and split-half reliability estimates are referred to as measures of ____________.
1. true scores; error scores
2. internal consistency; stability
3. inter-scorer reliability; consistency
4. stability; internal consistency
Which of the following is usually minimized when using split-half reliability estimates as compared with test-retest or parallel forms reliability estimates?
1. time and expense
2. reliability and validity
3. reliability only
4. none are correct
Which of the following factors may influence a split-half reliability estimate?
1. fatigue
2. anxiety
3. item difficulty
4. all are correct
For a heterogeneous test, measures of internal consistency reliability will tend to be ____________ compared with other methods of estimating reliability.
1. higher
2. lower
3. very similar or higher
4. not effected by heterogeneous items
Error variance for measures of inter-item consistency comes from:
1. fatigue
2. motivation
3. a test-taker practice effect
4. heterogeneity of the content
If items from a test are measuring the same trait, then estimates of reliability yielded from split-half methods will typically be ____________ compared with KR-20.
1. higher
2. lower
3. similar
4. exactly the same
Which of the following is NOT an acceptable way to divide a test when using the split-half method of reliability?
1. randomly assign items to each half
2. assign odd-numbered items to one half and even-numbered items to the other half
3. assign the first half of the items to one half and the second half of the items to the other half
4. assign easy items to one half and difficult items to the other half
Which is NOT an assumption that should be met in order to use KR-21?
1. items should be dichotomous
2. items should be of equal difficulty
3. items should be homogeneous
4. items should be scored by the same scorer
Which is a synonym for inter-scorer reliability?
1. inter-judge reliability
2. observer reliability
3. inter-rater reliability
4. all are correct
Which BEST conveys the meaning of an inter-scorer reliability estimate of .90?
1. 90% of the scores obtained are reliable
2. 90% of the variance in the scores assigned by the scorers was attributed to true differences and 10% to error
3. 10% of the variance in the scores assigned by the scorers was attributed to true differences and 90% to error
4. the test is stable
Which type(s) of reliability estimates would be most appropriate for a measure of heart rate?
1. test-retest
2. alternate form
3. inter-rater
4. all are correct
If a time limit is long enough to allow test-takers to attempt all items, and if some items are so difficult that no test-taker is able to obtain a perfect score, then the test is referred to as a ____________ test.
1. speed
2. power
3. reliable
4. valid
Interpretations of criterion-referenced tests are typically made with respect to:
1. the total number of items the examinee responded to
2. the material the examinee evidenced mastery of
3. a comparison of the examinee's performance with that of others who took the test
4. the total number of items the examinee passed or failed
The statement "68% of the scores for a particular test fall between 58 and 61" is:
1. a confidence interval
2. the reliability of the test
3. the validity of the test
4. all are correct
The standard error of measurement of a particular test of anxiety is 8. A student earns a score of 60. What is the confidence interval for this test score at the 95% confidence level?
1. 52-68
2. 40-68
3. 44-76
4. 36-84
As the confidence interval increases, the range of scores a single test score is likely to fall into:
1. decreases
2. increases
3. remains the same
4. first increases, then decreases
If the standard deviations of two tests are identical, but the reliability is lower for Test A as compared to Test B, then the standard error of measurement will be ____________ for Test A as compared with Test B.
1. higher
2. lower
3. the same
4. lower or the same

Chapter 5: Reliability

Choose the correct answer for each question.